Remote track fault detection system and method based on big data

Abstract

The invention discloses a remote track fault detection system and method based on big data, and relates to the technical field of big data, and the method comprises the steps: carrying out the multi-segment segmentation of a target track, and collecting the static track image data and the dynamic track image data; on the basis of image acquisition data of each section of track, constructing a track array point motion model to perform array point mapping decomposition on the static track, simulating array point motion on the dynamic track, and performing deformation elasticity index analysis on each section of track on the basis of the track array point motion model; performing array point classification on each section of track based on the array point loss interval by combining the deformation elasticity index analysis data of each section of track, and analyzing the deformation loss value of the corresponding section of track; carrying out comprehensive track loss evaluation by combining the deformation loss value of each section of the track on the target track; and outputting the segmented track array point model of the target track through a visual port, marking the deformation loss value corresponding to each segment of track, and feeding back the comprehensive loss condition of the target track.

Description

Technical Field

[0001] The present invention relates to the field of big data, and in particular to a remote track fault detection system and method based on big data. Background Art

[0002] Remote rail fault detection based on big data refers to the use of big data technology to integrate rail transit infrastructure detection data, and through data mining, analysis and prediction, to perform fault detection and status prediction on key parts of the track, so as to improve equipment life management and safety monitoring capabilities; the current rail transit covers a wide range, involves many cities, and has a high degree of integration, so rail fault detection is an important link in maintaining the long-term development of rail transit; in the current environment, the detection methods for tracks are still relatively backward, and often require regular manual inspections at fixed points or post-fault maintenance, which poses a great obstacle to track operation; and due to the inherent properties of the track, its latent faults are often difficult to be detected manually, so most track faults cannot be discovered in advance, resulting in abnormal track wear. Summary of the invention

[0003] The purpose of the present invention is to provide a remote track fault detection system and method based on big data to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solutions: A remote rail fault detection method based on big data, the method comprising the following steps: S100, collecting images of the track through image acquisition; dividing the target track into multiple segments, monitoring the status of each track segment, and collecting static track image data and dynamic track image data; S200, based on the image data of each track segment of the target track, the static track is decomposed by constructing a track lattice point motion model, and the lattice point motion on the dynamic track is simulated, and the deformation elastic index of each track segment is analyzed based on the track lattice point motion model; S300, combining deformation elasticity index analysis data of each track section, classifying each track section based on the lattice point loss interval, and analyzing the deformation loss value of the corresponding segmented track in combination with the corresponding track lattice point classification data; and performing a comprehensive track loss assessment in combination with the deformation loss value of each track section on the target track; S400: Output the segmented track lattice point model of the target track through the visual port, mark the deformation loss value of each corresponding track segment, and feedback the comprehensive loss situation of the target track.

[0005] The specific steps of S100 for collecting images of the track through image collection, dividing the target track into multiple segments, monitoring the status of each segment of the track, and collecting static track image data and dynamic track image data are as follows: S101, dividing the target track into equal sections, obtaining multiple sections of the target track after the cutting, and assigning labels to each section of the track; S102. By remotely monitoring the status of each track section corresponding to the target track, static track image acquisition is performed based on the static track; and motion state periodic image acquisition is performed based on the dynamic track; wherein the static track is a track in a state where no vehicle passes; and the dynamic track is a track in a state where a vehicle passes; wherein the period is the time length between a static track state and an adjacent dynamic track state.

[0006] The S200 collects image data of each track segment of the target track, decomposes the static track by constructing a track lattice point motion model, and simulates the lattice point motion on the dynamic track. The specific steps of performing deformation elastic index analysis on each track segment based on the track lattice point motion model are as follows: S201, retrieve the continuous state image data corresponding to each track segment of the target track within the cycle; construct a track lattice point motion model, input the collected image data into the model, and obtain the lattice point distribution map corresponding to each track segment in the static state and the lattice point motion simulation map in the dynamic state; establish a plane coordinate system, perform data dimension reduction on the static image data corresponding to each track segment, obtain the discretized lattice point data of the corresponding segmented track and map it to the coordinate system, and construct the static lattice point distribution map of the corresponding track; based on the static lattice point distribution map of the corresponding segmented track, simulate the continuous motion trajectory map of each lattice point by combining the continuous dynamic data of the corresponding segmented track with the coordinate system; S202, analyzing the periodic static deformation elastic index of the corresponding segmented track in the static lattice distribution diagram of each segment track, and the calculation formula is: ; Wherein, FJ(n,m) is the periodic static deformation elastic index of the lattice point numbered m on the segmented track corresponding to label n; m is the number of lattice points on the segmented track corresponding to label n; S[C(m):C(m0)] is the coordinate distance between the lattice point numbered m on the segmented track corresponding to label n and the initial position of the corresponding lattice point; C(m) is the coordinate of the lattice point numbered m on the segmented track corresponding to label n; C(m0) is the coordinate of the initial position of the lattice point on the segmented track corresponding to label n S[C(m):C(m0)] max is the maximum value of the coordinate distance between the lattice point numbered m on the segmented track corresponding to label n and the initial position of the corresponding lattice point; In the continuous motion trajectory diagram of the dynamic array points of each track segment, the periodic dynamic deformation elastic index of the corresponding segmented track is analyzed, and the calculation formula is: ; Wherein, FD(n,m) is the dynamic deformation elasticity index of the lattice point m of the segmented track with label n; S[C(m,t):C(m,t-1))] is the distance between the coordinates of the lattice point m of the segmented track with label n at time t and the coordinates at time t-1 in the motion trajectory; C(m,t) is the coordinates of the lattice point m of the segmented track with label n at time t in the motion trajectory; C(m,t-1) is the coordinates of the lattice point m of the segmented track with label n at time t-1 in the motion trajectory; x is the number of time points of the motion trajectory of each lattice point on the segmented track with label n; S[C(m,tx):C(m,t1)] max The maximum value of the corresponding coordinate distance between two time points of the moving time interval [t1, tx] of the lattice point numbered m in the segmented track corresponding to the label n; when t is 1, the previous moment is 0, which represents the coordinate position of the lattice point in the static state.

[0007] The S300 combines the deformation elasticity index analysis data of each track section, classifies the points of each track section based on the point loss interval, and analyzes the deformation loss value of the corresponding segmented track in combination with the corresponding track point classification data; the specific steps of conducting a comprehensive track loss assessment in combination with the deformation loss value of each track section on the target track are as follows: S301, by calling the track deformation loss comparison parameter list, determine the deformation loss interval of each static track array point on each track section and the deformation loss interval of each dynamic track array point motion trajectory, and classify the array points, and call the loss influence parameters of the corresponding loss interval; perform periodic static deformation loss analysis and periodic dynamic deformation loss analysis on each track section respectively, and the calculation formula is: ; ; Where SJ(n) is the periodic static deformation loss value of the segmented track with label n; α(k) is the loss influence parameter of the deformation loss interval with number k; FJ(n,m:k) is the static deformation elasticity index of the static array point belonging to the deformation loss interval with number k on the segmented track with label n; FJ max is the maximum allowable static deformation elasticity index of the static array point; SD(n) is the periodic dynamic deformation loss value of the segmented track corresponding to label n; FD(n,m:k) is the dynamic deformation elasticity index of the dynamic array point belonging to the deformation loss interval numbered k on the segmented track corresponding to label n; FD maxis the maximum allowable dynamic deformation elastic index of the dynamic array point; combined with the periodic static deformation loss analysis and periodic dynamic deformation loss analysis data of the corresponding segmented track, a periodic comprehensive loss analysis is performed on the segmented track, and the calculation formula is: ; Among them, SZ(J) is the periodic comprehensive loss value of the segmented track corresponding to label n; g(J) and g(D) are the static deformation loss influence weight and dynamic deformation loss influence weight respectively; T is the number of periodic time points; S302, based on the periodic comprehensive loss analysis data of each section of the target track, a periodic comprehensive loss analysis is performed on the target track, and the calculation formula is: ; Among them, SG is the periodic comprehensive loss degree of the target orbit; SZ(n)max and SZ(n)min correspond to the maximum and minimum periodic comprehensive loss values ​​in each segment of the target orbit, respectively; Based on the periodic comprehensive loss degree analysis data of the target orbit, the loss warning threshold SG(p) is set to judge the loss situation of the target orbit in the current period; if SG<SG(p), it is judged that the loss of the target orbit in the current period is normal; if SG≥SG(p), it is judged that the loss of the target orbit in the current period is abnormal.

[0008] The S400 outputs the segmented track lattice point model of the target track through the visual port, and marks the deformation loss value of each corresponding track segment. The specific steps of feeding back the comprehensive loss situation of the target track are as follows: S401, using a display port to visualize and output the segmentation of the target track and the track lattice model; S402, marking and displaying the deformation loss value of each track section in the target track array point model, outputting the comprehensive loss status of the target track, and issuing a port warning based on the abnormal loss situation.

[0009] A remote track fault detection system based on big data, the system comprising a track data acquisition module, a lattice model building module, a track comprehensive loss assessment module and a data feedback module; The track data acquisition module acquires images of the track through image acquisition; divides the target track into multiple sections, monitors the status of each track section, and acquires static track image data and dynamic track image data; the lattice point model construction module decomposes the static track by constructing a track lattice point motion model based on the image acquisition data of each track section of the target track, simulates the lattice point motion on the dynamic track, and analyzes the deformation elasticity index of each track section based on the track lattice point motion model; the track comprehensive loss assessment module classifies the lattice points of each track section based on the lattice point loss interval in combination with the deformation elasticity index analysis data of each track section, and analyzes the deformation loss value of the corresponding segmented track in combination with the corresponding lattice point classification data of each track; and performs a comprehensive track loss assessment in combination with the deformation loss value of each track section on the target track; the data feedback module outputs the segmented track lattice point model of the target track through the visual port, marks the deformation loss value of each corresponding track section, and feeds back the comprehensive loss of the target track.

[0010] The track data acquisition module includes a track simulation segmentation unit and a track image data acquisition unit; The track simulation segmentation unit segments the target track into equal-sized segments, obtains multiple track segments after the target track is cut, and assigns labels to each track segment; The track image data acquisition unit remotely monitors the status of each track section corresponding to the target track, performs static track image acquisition based on the static track, and performs motion state periodic image acquisition based on the dynamic track.

[0011] The lattice model building module includes a track lattice model building unit and a track deformation elasticity analysis unit; The track lattice point model construction unit retrieves the continuous state image data corresponding to each track segment of the target track within the cycle; constructs a track lattice point motion model, and inputs the collected image data into the model to obtain the lattice point distribution diagram corresponding to each track segment in the static state and the lattice point motion simulation diagram in the dynamic state; establishes a plane coordinate system, performs data dimension reduction on the static image data corresponding to each track segment, obtains the discretized lattice point data of the corresponding segmented track and maps it to the coordinate system, and constructs the static lattice point distribution diagram of the corresponding track; based on the static lattice point distribution diagram of the corresponding segmented track, simulates the continuous motion trajectory diagram of each lattice point by combining the continuous dynamic data of the corresponding segmented track with the coordinate system; The track deformation elasticity analysis unit analyzes the periodic static deformation elasticity index of the corresponding segmented track in the static lattice point distribution diagram of each segmented track; and analyzes the periodic dynamic deformation elasticity index of the corresponding segmented track in the continuous motion trajectory diagram of the dynamic lattice points of each segmented track.

[0012] The track comprehensive loss assessment module includes a track state deformation loss analysis unit and a track comprehensive loss assessment unit; The track state deformation loss analysis unit retrieves the track deformation loss comparison parameter list, determines the deformation loss interval of each static track lattice point on each track segment and the deformation loss interval of each dynamic track lattice point motion trajectory, and classifies the lattice points, retrieves the loss influence parameters of the corresponding loss interval; performs periodic static deformation loss analysis and periodic dynamic deformation loss analysis on each track segment respectively; combines the periodic static deformation loss analysis and periodic dynamic deformation loss analysis data of the corresponding segmented track, and performs periodic comprehensive loss analysis on the segmented track; The track comprehensive loss assessment unit performs a periodic comprehensive loss analysis on the target track based on the periodic comprehensive loss analysis data of each section of the target track; and sets a loss warning threshold based on the periodic comprehensive loss degree analysis data of the target track to judge the loss situation of the target track in the current period.

[0013] The data feedback module includes a data visualization unit and an abnormal loss warning unit; The data visualization unit uses a display port to visualize and output the segmentation of the target track and the track lattice point model; The abnormal loss warning unit marks and displays the deformation loss value of each section of the track in the target track lattice model, outputs the comprehensive loss status of the target track, and issues a port warning based on the abnormal loss situation.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention realizes the monitoring of periodic track loss and abnormal warning based on track image acquisition, model construction and comprehensive analysis of track status data; realizes multi-threaded small layout data analysis through segmented track analysis, realizes pin-point simulation of periodic static image and dynamic image data by using track pin-point model, and analyzes the state deformation elasticity index of target track based on the model; obtains the deformation change of periodic target track, and comprehensively analyzes the loss of segmented track in combination with dynamic and static states; finally, comprehensively analyzes the loss of each segmented track; the present invention realizes abnormal detection of periodic target track by using remote thread periodic dynamic and static state track data analysis, improves the limitations of traditional manual detection, and improves the degree of autonomy of track fault detection and fault mining capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a structural schematic diagram of a remote track fault detection system based on big data of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Example: Figure 1 As shown, the present invention provides a technical solution: A remote rail fault detection method based on big data, the method comprising the following steps: S100, collecting images of the track through image acquisition; dividing the target track into multiple segments, monitoring the status of each track segment, and collecting static track image data and dynamic track image data; S200, based on the image data of each track segment of the target track, the static track is decomposed by constructing a track lattice point motion model, and the lattice point motion on the dynamic track is simulated, and the deformation elastic index of each track segment is analyzed based on the track lattice point motion model; S300, combining deformation elasticity index analysis data of each track section, classifying each track section based on the lattice point loss interval, and analyzing the deformation loss value of the corresponding segmented track in combination with the corresponding track lattice point classification data; and performing a comprehensive track loss assessment in combination with the deformation loss value of each track section on the target track; S400: Output the segmented track lattice point model of the target track through the visual port, mark the deformation loss value of each corresponding track segment, and feedback the comprehensive loss situation of the target track.

[0018] The specific steps of S100 for collecting images of the track through image collection, dividing the target track into multiple segments, monitoring the status of each segment of the track, and collecting static track image data and dynamic track image data are as follows: S101, dividing the target track into equal sections, obtaining multiple sections of the target track after the cutting, and assigning labels to each section of the track; S102, by remotely monitoring the status of each track section corresponding to the target track, static track image acquisition is performed based on the static track; and motion state periodic image acquisition is performed based on the dynamic track.

[0019] The S200 collects image data of each track segment of the target track, decomposes the static track by constructing a track lattice point motion model, and simulates the lattice point motion on the dynamic track. The specific steps of performing deformation elastic index analysis on each track segment based on the track lattice point motion model are as follows: S201, retrieve the continuous state image data corresponding to each track segment of the target track within the cycle; construct a track lattice point motion model, input the collected image data into the model, and obtain the lattice point distribution map corresponding to each track segment in the static state and the lattice point motion simulation map in the dynamic state; establish a plane coordinate system, perform data dimension reduction on the static image data corresponding to each track segment, obtain the discretized lattice point data of the corresponding segmented track and map it to the coordinate system, and construct the static lattice point distribution map of the corresponding track; based on the static lattice point distribution map of the corresponding segmented track, simulate the continuous motion trajectory map of each lattice point by combining the continuous dynamic data of the corresponding segmented track with the coordinate system; S202, analyzing the periodic static deformation elastic index of the corresponding segmented track in the static lattice distribution diagram of each segment track, and the calculation formula is: ; Wherein, FJ(n,m) is the periodic static deformation elastic index of the lattice point numbered m on the segmented track corresponding to label n; m is the number of lattice points on the segmented track corresponding to label n; S[C(m):C(m0)] is the coordinate distance between the lattice point numbered m on the segmented track corresponding to label n and the initial position of the corresponding lattice point; C(m) is the coordinate of the lattice point numbered m on the segmented track corresponding to label n; C(m0) is the coordinate of the initial position of the lattice point on the segmented track corresponding to label n S[C(m):C(m0)] max is the maximum value of the coordinate distance between the lattice point numbered m on the segmented track corresponding to label n and the initial position of the corresponding lattice point; In the continuous motion trajectory diagram of the dynamic array points of each track segment, the periodic dynamic deformation elastic index of the corresponding segmented track is analyzed, and the calculation formula is: ; Wherein, FD(n,m) is the dynamic deformation elasticity index of the lattice point m of the segmented track with label n; S[C(m,t):C(m,t-1))] is the distance between the coordinates of the lattice point m of the segmented track with label n at time t and the coordinates at time t-1 in the motion trajectory; C(m,t) is the coordinates of the lattice point m of the segmented track with label n at time t in the motion trajectory; C(m,t-1) is the coordinates of the lattice point m of the segmented track with label n at time t-1 in the motion trajectory; x is the number of time points of the motion trajectory of each lattice point on the segmented track with label n; S[C(m,tx):C(m,t1)] max The maximum value of the corresponding coordinate distance between two points in time on the moving time interval [t1, tx] for the segmented track number m lattice point with the corresponding label n.

[0020] The S300 combines the deformation elasticity index analysis data of each track section, classifies the points of each track section based on the point loss interval, and analyzes the deformation loss value of the corresponding segmented track in combination with the corresponding track point classification data; the specific steps of conducting a comprehensive track loss assessment in combination with the deformation loss value of each track section on the target track are as follows: S301, by calling the track deformation loss comparison parameter list, determine the deformation loss interval of each static track array point on each track section and the deformation loss interval of each dynamic track array point motion trajectory, and classify the array points, and call the loss influence parameters of the corresponding loss interval; perform periodic static deformation loss analysis and periodic dynamic deformation loss analysis on each track section respectively, and the calculation formula is: ; ; Where SJ(n) is the periodic static deformation loss value of the segmented track with label n; α(k) is the loss influence parameter of the deformation loss interval with number k; FJ(n,m:k) is the static deformation elasticity index of the static array point belonging to the deformation loss interval with number k on the segmented track with label n; FJ max is the maximum allowable static deformation elasticity index of the static array point; SD(n) is the periodic dynamic deformation loss value of the segmented track corresponding to label n; FD(n,m:k) is the dynamic deformation elasticity index of the dynamic array point belonging to the deformation loss interval numbered k on the segmented track corresponding to label n; FD max is the maximum allowable dynamic deformation elastic index of the dynamic array point; combined with the periodic static deformation loss analysis and periodic dynamic deformation loss analysis data of the corresponding segmented track, a periodic comprehensive loss analysis is performed on the segmented track, and the calculation formula is: ; Among them, SZ(J) is the periodic comprehensive loss value of the segmented track corresponding to label n; g(J) and g(D) are the static deformation loss influence weight and dynamic deformation loss influence weight respectively; T is the number of periodic time points; S302, based on the periodic comprehensive loss analysis data of each section of the target track, a periodic comprehensive loss analysis is performed on the target track, and the calculation formula is: ; Among them, SG is the periodic comprehensive loss degree of the target orbit; SZ(n)max and SZ(n)min correspond to the maximum and minimum periodic comprehensive loss values ​​in each segment of the target orbit, respectively; Based on the periodic comprehensive loss degree analysis data of the target orbit, the loss warning threshold SG(p) is set to judge the loss situation of the target orbit in the current period; if SG<SG(p), it is judged that the loss of the target orbit in the current period is normal; if SG≥SG(p), it is judged that the loss of the target orbit in the current period is abnormal.

[0021] The S400 outputs the segmented track lattice point model of the target track through the visual port, and marks the deformation loss value of each corresponding track segment. The specific steps of feeding back the comprehensive loss situation of the target track are as follows: S401, using a display port to visualize and output the segmentation of the target track and the track lattice model; S402, marking and displaying the deformation loss value of each track section in the target track array point model, outputting the comprehensive loss status of the target track, and issuing a port warning based on the abnormal loss situation.

[0022] A remote track fault detection system based on big data, the system comprising a track data acquisition module, a lattice model building module, a track comprehensive loss assessment module and a data feedback module; The track data acquisition module acquires images of the track through image acquisition; divides the target track into multiple sections, monitors the status of each track section, and acquires static track image data and dynamic track image data; the lattice point model construction module decomposes the static track by constructing a track lattice point motion model based on the image acquisition data of each track section of the target track, simulates the lattice point motion on the dynamic track, and analyzes the deformation elasticity index of each track section based on the track lattice point motion model; the track comprehensive loss assessment module classifies the lattice points of each track section based on the lattice point loss interval in combination with the deformation elasticity index analysis data of each track section, and analyzes the deformation loss value of the corresponding segmented track in combination with the corresponding lattice point classification data of each track; and performs a comprehensive track loss assessment in combination with the deformation loss value of each track section on the target track; the data feedback module outputs the segmented track lattice point model of the target track through the visual port, marks the deformation loss value of each corresponding track section, and feeds back the comprehensive loss of the target track.

[0023] The track data acquisition module includes a track simulation segmentation unit and a track image data acquisition unit; The track simulation segmentation unit segments the target track into equal-sized segments, obtains multiple track segments after the target track is cut, and assigns labels to each track segment; The track image data acquisition unit remotely monitors the status of each track section corresponding to the target track, performs static track image acquisition based on the static track, and performs motion state periodic image acquisition based on the dynamic track.

[0024] The lattice model building module includes a track lattice model building unit and a track deformation elasticity analysis unit; The track lattice point model construction unit retrieves the continuous state image data corresponding to each track segment of the target track within the cycle; constructs a track lattice point motion model, and inputs the collected image data into the model to obtain the lattice point distribution diagram corresponding to each track segment in the static state and the lattice point motion simulation diagram in the dynamic state; establishes a plane coordinate system, performs data dimension reduction on the static image data corresponding to each track segment, obtains the discretized lattice point data of the corresponding segmented track and maps it to the coordinate system, and constructs the static lattice point distribution diagram of the corresponding track; based on the static lattice point distribution diagram of the corresponding segmented track, simulates the continuous motion trajectory diagram of each lattice point by combining the continuous dynamic data of the corresponding segmented track with the coordinate system; The track deformation elasticity analysis unit analyzes the periodic static deformation elasticity index of the corresponding segmented track in the static lattice point distribution diagram of each segmented track; and analyzes the periodic dynamic deformation elasticity index of the corresponding segmented track in the continuous motion trajectory diagram of the dynamic lattice points of each segmented track.

[0025] The track comprehensive loss assessment module includes a track state deformation loss analysis unit and a track comprehensive loss assessment unit; The track state deformation loss analysis unit retrieves the track deformation loss comparison parameter list, determines the deformation loss interval of each static track lattice point on each track segment and the deformation loss interval of each dynamic track lattice point motion trajectory, and classifies the lattice points, retrieves the loss influence parameters of the corresponding loss interval; performs periodic static deformation loss analysis and periodic dynamic deformation loss analysis on each track segment respectively; combines the periodic static deformation loss analysis and periodic dynamic deformation loss analysis data of the corresponding segmented track, and performs periodic comprehensive loss analysis on the segmented track; The track comprehensive loss assessment unit performs a periodic comprehensive loss analysis on the target track based on the periodic comprehensive loss analysis data of each section of the target track; and sets a loss warning threshold based on the periodic comprehensive loss degree analysis data of the target track to judge the loss situation of the target track in the current period.

[0026] The data feedback module includes a data visualization unit and an abnormal loss warning unit; The data visualization unit uses a display port to visualize and output the segmentation of the target track and the track lattice point model; The abnormal loss warning unit marks and displays the deformation loss value of each section of the track in the target track lattice model, outputs the comprehensive loss status of the target track, and issues a port warning based on the abnormal loss situation; In example:

[0027] A certain track supervision unit has introduced the remote track fault detection system based on big data to realize remote monitoring and fault detection of the target track; it divides the target track into equal-specification segments, obtains multiple segments of the target track after cutting, and assigns labels to each segment of the track; it remotely monitors the status of each segment of the target track, performs static track image acquisition based on the static track; and performs motion state periodic image acquisition based on the dynamic track; Retrieve the corresponding continuous state image data of each segment of the target track within the period; construct a track lattice point motion model, input the collected image data into the model, and obtain the lattice point distribution map of each segment of the track in the static state and the lattice point motion simulation map in the dynamic state; establish a plane coordinate system, perform data dimension reduction on the static image data of each segment of the track, obtain the discretized lattice point data of the corresponding segmented track and map it to the coordinate system, and construct the static lattice point distribution map of the corresponding track; based on the static lattice point distribution map of the corresponding segmented track, simulate the continuous motion trajectory map of each lattice point by combining the continuous dynamic data of the corresponding segmented track with the coordinate system; analyze the periodic static deformation elasticity index of the corresponding segmented track in the static lattice point distribution map of each segment of the track, and the calculation formula is: ; In the continuous motion trajectory diagram of the dynamic array points of each track segment, the periodic dynamic deformation elastic index of the corresponding segmented track is analyzed, and the calculation formula is: ; By calling up the track deformation loss comparison parameter list, the deformation loss interval of each static track array point on each track section and the deformation loss interval of each dynamic track array point motion trajectory are determined to classify the array points, and the loss influence parameters of the corresponding loss interval are called up; the periodic static deformation loss analysis and periodic dynamic deformation loss analysis are performed on each track section respectively, and the calculation formula is: ; ; Combining the periodic static deformation loss analysis and periodic dynamic deformation loss analysis data of the corresponding segmented track, a periodic comprehensive loss analysis of the segmented track is performed, and the calculation formula is: ; Based on the periodic comprehensive loss analysis data of each section of the target orbit, the periodic comprehensive loss analysis of the target orbit is performed, and the calculation formula is: ; Based on the periodic comprehensive loss degree analysis data of the target track, a loss warning threshold SG(p) is set to judge the loss of the target track in the current period; if SG<SG(p), the loss of the target track in the current period is judged to be normal; if SG≥SG(p), the loss of the target track in the current period is judged to be abnormal; The display port is used to visualize the segmentation of the target track and the track lattice model; the deformation loss value of each track segment is marked and displayed in the target track lattice model, and the comprehensive loss status of the target track is output, and a port warning is issued based on abnormal loss conditions.

[0028] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A remote rail fault detection method based on big data, characterized in that: The method comprises the following steps: S100, collecting images of the track through image acquisition; dividing the target track into multiple segments, monitoring the status of each track segment, and collecting static track image data and dynamic track image data; S200, based on the image data of each track segment of the target track, the static track is decomposed by constructing a track lattice point motion model, and the lattice point motion on the dynamic track is simulated, and the deformation elastic index of each track segment is analyzed based on the track lattice point motion model; S300, combining deformation elasticity index analysis data of each track section, classifying each track section based on the lattice point loss interval, and analyzing the deformation loss value of the corresponding segmented track in combination with the corresponding track lattice point classification data; and performing a comprehensive track loss assessment in combination with the deformation loss value of each track section on the target track; S400: Output the segmented track lattice point model of the target track through the visual port, mark the deformation loss value of each corresponding track segment, and feedback the comprehensive loss situation of the target track.

2. A remote rail fault detection method based on big data according to claim 1, characterized in that: The specific steps of S100 for collecting images of the track through image collection, dividing the target track into multiple segments, monitoring the status of each segment of the track, and collecting static track image data and dynamic track image data are as follows: S101, dividing the target track into equal sections, obtaining multiple sections of the target track after the cutting, and assigning labels to each section of the track; S102, by remotely monitoring the status of each track section corresponding to the target track, static track image acquisition is performed based on the static track; and motion state periodic image acquisition is performed based on the dynamic track.

3. A remote rail fault detection method based on big data according to claim 2, characterized in that: The S200 collects image data of each track segment of the target track, decomposes the static track by constructing a track lattice point motion model, and simulates the lattice point motion on the dynamic track. The specific steps of performing deformation elastic index analysis on each track segment based on the track lattice point motion model are as follows: S201, retrieve the continuous state image data corresponding to each track segment of the target track within the cycle; construct a track lattice point motion model, input the collected image data into the model, and obtain the lattice point distribution map corresponding to each track segment in the static state and the lattice point motion simulation map in the dynamic state; establish a plane coordinate system, perform data dimension reduction on the static image data corresponding to each track segment, obtain the discretized lattice point data of the corresponding segmented track and map it to the coordinate system, and construct the static lattice point distribution map of the corresponding track; based on the static lattice point distribution map of the corresponding segmented track, simulate the continuous motion trajectory map of each lattice point by combining the continuous dynamic data of the corresponding segmented track with the coordinate system; S202, analyzing the periodic static deformation elastic index of the corresponding segmented track in the static lattice distribution diagram of each segment track, and the calculation formula is: ； Wherein, FJ(n,m) is the periodic static deformation elastic index of the numbered m lattice point on the segmented track corresponding to label n; m is the number of lattice points on the segmented track corresponding to label n; S[C(m):C(m0)] is the coordinate distance between the numbered m lattice point on the segmented track corresponding to label n and the initial position of the corresponding lattice point; C(m) is the coordinate of the numbered m lattice point on the segmented track corresponding to label n; C(m0) is the coordinate of the initial position of the lattice point on the segmented track corresponding to label n S[C(m):C(m0)] max is the maximum value of the coordinate distance between the lattice point numbered m on the segmented track corresponding to label n and the initial position of the corresponding lattice point; In the continuous motion trajectory diagram of the dynamic array points of each track segment, the periodic dynamic deformation elastic index of the corresponding segmented track is analyzed, and the calculation formula is: ； Wherein, FD(n,m) is the dynamic deformation elasticity index of the lattice point m of the segmented track with label n; S[C(m,t):C(m,t-1))] is the distance between the coordinates of the lattice point m of the segmented track with label n at time t and the coordinates at time t-1 in the motion trajectory; C(m,t) is the coordinates of the lattice point m of the segmented track with label n at time t in the motion trajectory; C(m,t-1) is the coordinates of the lattice point m of the segmented track with label n at time t-1 in the motion trajectory; x is the number of time points of the motion trajectory of each lattice point on the segmented track with label n; S[C(m,tx):C(m,t1)] max The maximum value of the corresponding coordinate distance between two points in time on the moving time interval [t1, tx] for the segmented track number m lattice point with the corresponding label n.

4. The method for remote rail fault detection based on big data according to claim 3 is characterized in that: The S300 combines the deformation elasticity index analysis data of each track section, classifies the points of each track section based on the point loss interval, and analyzes the deformation loss value of the corresponding segmented track in combination with the corresponding track point classification data; the specific steps of conducting a comprehensive track loss assessment in combination with the deformation loss value of each track section on the target track are as follows: S301, by calling the track deformation loss comparison parameter list, determine the deformation loss interval of each static track array point on each track section and the deformation loss interval of each dynamic track array point motion trajectory, and classify the array points, and call the loss influence parameters of the corresponding loss interval; perform periodic static deformation loss analysis and periodic dynamic deformation loss analysis on each track section respectively, and the calculation formula is: ； ； Where SJ(n) is the periodic static deformation loss value of the segmented track with label n; α(k) is the loss influence parameter of the deformation loss interval with number k; FJ(n,m:k) is the static deformation elasticity index of the static array point belonging to the deformation loss interval with number k on the segmented track with label n; FJ max is the maximum allowable static deformation elasticity index of the static array point; SD(n) is the periodic dynamic deformation loss value of the segmented track corresponding to label n; FD(n,m:k) is the dynamic deformation elasticity index of the dynamic array point belonging to the deformation loss interval numbered k on the segmented track corresponding to label n; FD max is the maximum allowable dynamic deformation elastic index of the dynamic array point; combined with the periodic static deformation loss analysis and periodic dynamic deformation loss analysis data of the corresponding segmented track, a periodic comprehensive loss analysis is performed on the segmented track, and the calculation formula is: ； Among them, SZ(J) is the periodic comprehensive loss value of the segmented track corresponding to label n; g(J) and g(D) are the static deformation loss influence weight and dynamic deformation loss influence weight respectively; T is the number of periodic time points; S302, based on the periodic comprehensive loss analysis data of each section of the target track, a periodic comprehensive loss analysis is performed on the target track, and the calculation formula is: ； Among them, SG is the periodic comprehensive loss degree of the target orbit; SZ(n)max and SZ(n)min correspond to the maximum and minimum periodic comprehensive loss values ​​in each segment of the target orbit, respectively; Based on the periodic comprehensive loss degree analysis data of the target orbit, the loss warning threshold SG(p) is set to judge the loss situation of the target orbit in the current period; if SG<SG(p), it is judged that the loss of the target orbit in the current period is normal; if SG≥SG(p), it is judged that the loss of the target orbit in the current period is abnormal.

5. A remote rail fault detection method based on big data according to claim 4, characterized in that: The S400 outputs the segmented track lattice point model of the target track through the visual port, and marks the deformation loss value of each corresponding track segment. The specific steps of feeding back the comprehensive loss situation of the target track are as follows: S401, using a display port to visualize and output the segmentation of the target track and the track lattice model; S402, marking and displaying the deformation loss value of each track section in the target track array point model, outputting the comprehensive loss status of the target track, and issuing a port warning based on the abnormal loss situation.

6. A remote track fault detection system based on big data, characterized in that: The system includes a track data acquisition module, a lattice model building module, a track comprehensive loss evaluation module and a data feedback module; The track data acquisition module acquires images of the track through image acquisition; divides the target track into multiple sections, monitors the status of each section of the track, and acquires static track image data and dynamic track image data; the lattice point model construction module, based on the track image acquisition data of each section of the target track, constructs a track lattice point motion model to perform lattice point mapping decomposition on the static track, simulates the lattice point motion on the dynamic track, and performs deformation elasticity index analysis on each section of the track based on the track lattice point motion model; The track comprehensive loss assessment module combines the deformation elasticity index analysis data of each track section, classifies the points of each track section based on the point loss interval, and analyzes the deformation loss value of the corresponding segmented track in combination with the corresponding track point classification data; A comprehensive track loss assessment is performed in combination with the deformation loss values ​​of each track section on the target track; the data feedback module outputs the segmented track lattice model of the target track through a visual port, and marks the deformation loss values ​​of each corresponding track section, and feeds back the comprehensive loss situation of the target track.

7. A remote rail fault detection system based on big data according to claim 6, characterized in that: The track data acquisition module includes a track simulation segmentation unit and a track image data acquisition unit; The track simulation segmentation unit segments the target track into equal-sized segments, obtains multiple track segments after the target track is cut, and assigns labels to each track segment; The track image data acquisition unit remotely monitors the status of each track section corresponding to the target track, performs static track image acquisition based on the static track, and performs motion state periodic image acquisition based on the dynamic track.

8. The remote rail fault detection system based on big data according to claim 7 is characterized in that: The lattice model building module includes a track lattice model building unit and a track deformation elasticity analysis unit; The track lattice point model construction unit retrieves the continuous state image data corresponding to each track segment of the target track within the period; By constructing a track lattice point motion model and inputting the collected image data into the model, the lattice point distribution diagram of each track section in a static state and the lattice point motion simulation diagram in a dynamic state are obtained; by establishing a plane coordinate system, the static image data of each track section is subjected to data dimension reduction, the discretized lattice point data of the corresponding segmented track is obtained and mapped to the coordinate system, and the static lattice point distribution diagram of the corresponding track is constructed; based on the static lattice point distribution diagram of the corresponding segmented track, the continuous motion trajectory diagram of each lattice point is simulated by combining the continuous dynamic data of the corresponding segmented track with the coordinate system; The track deformation elasticity analysis unit analyzes the periodic static deformation elasticity index of the corresponding segmented track in the static lattice point distribution diagram of each segment track; In the continuous motion trajectory diagram of the dynamic array points of each track segment, the periodic dynamic deformation elastic index of the corresponding segmented track is analyzed.

9. A remote rail fault detection system based on big data according to claim 8, characterized in that: The track comprehensive loss assessment module includes a track state deformation loss analysis unit and a track comprehensive loss assessment unit; The track state deformation loss analysis unit retrieves the track deformation loss comparison parameter list, determines the deformation loss interval of each static track lattice point on each track segment and the deformation loss interval of each dynamic track lattice point motion trajectory, and classifies the lattice points, retrieves the loss influence parameters of the corresponding loss interval; performs periodic static deformation loss analysis and periodic dynamic deformation loss analysis on each track segment respectively; combines the periodic static deformation loss analysis and periodic dynamic deformation loss analysis data of the corresponding segmented track, and performs periodic comprehensive loss analysis on the segmented track; The track comprehensive loss assessment unit performs a periodic comprehensive loss analysis on the target track based on the periodic comprehensive loss analysis data of each section of the target track; and sets a loss warning threshold based on the periodic comprehensive loss degree analysis data of the target track to judge the loss situation of the target track in the current period.

10. A remote rail fault detection system based on big data according to claim 9, characterized in that: The data feedback module includes a data visualization unit and an abnormal loss warning unit; The data visualization unit uses a display port to visualize and output the segmentation of the target track and the track lattice point model; The abnormal loss warning unit marks and displays the deformation loss value of each section of the track in the target track lattice model, outputs the comprehensive loss status of the target track, and issues a port warning based on the abnormal loss situation.

Images